In continuing attempts to promote the psychological well-being of captive nonhuman primates, many institutional environmental enrichment plans include provisions for social housing. Pair-housing macaques in sqeeze-back cages provides a social environment while maintaining accessibility to the animals for research requirements. The objective of this study was to develop and evaluate materials and methods for safely and successfully pair-housing macaques which had previously been singly housed. Existing 4.3 square foot primate cages on racks were modified to facilitate pair-housing by cutting openings, which were fitted with removable wire grid partitions, in adjoining sides of each cage. Male and female rhesus and cynomolgus macaques ranging in age from less than one year to 22 years ( = 8 years) were used in this study. Potential pairs of the same species and sex were initially housed in adjoining cages with both partitions in place to permit visual and auditory non-contact communication. One partition was subsequently removed to permit limited direct physical contact through the remaining partition. Animals were observed for affiliative and/or aggressive behaviors and for establishment of dominant-subordinate relationships. Animals which did not exhibit marked aggressive behaviors were subsequently paired, following a minimum of 24 hours acclimation, by removal of the remaining partition. Animals were monitored continuously for the first 30 minutes after pairing and frequently during the first week. Of 25 attempted pairings, 8 pairs (32%) were considered incompatible because of aggression during the acclimation period. Fourteen of the 17 pairs subsequently formed (82%) were successful. One female pair was separated during the first week because of aggression, and two male pairs were separated because of aggression which became evident at 5 weeks and 3 months after pairing. In conclusion, macaques were safely and successfully pair-housed using the described